seqret reading from stdin with unkown type

[Peter Rice]

David Mathog wrote:
> 
> I don't understand why this one fails:
> 
> # ( echo "AGTTT" ; echo "ACGTTTTTGGT" ) \
>   | seqret -sequence=stdin  -out=gcg::stdout
> Reads and writes (returns) sequences
> Error: Unable to read sequence 'stdin'
>
> According the seqret blurb the program is supposed to try
> one format at a time until it finds something that works.  In
> the first case it seems not to be trying "plain".

Works for me (with 2.4.0).

Plain is not used by default, but "raw" is, which should accept
alphanumeric characters and 'whitespace', defined as
ajRegCompC("[^A-Za-z0-9 \t\n\r]") but reject anything else.

The trouble with "plain" (also called "text") is that it can accept
anything, usually with the wrong results. It silently removes non-sequence
characters from any input. As a default format, this a "A Bad Thing."
Turning it off lets us generate the error messages for bad formats.

So, the real questions is "why does your script fail with -sf raw" ? One
possibility is a problem with some of the other default formats not
cleaning up properly, which was fixed in 2.3.0.

Can you try with "seqret -debug -sf raw" and send the seqret.dbg file, just
to check.

I think all you need to do is upgrade to 2.4.0 and all should be well.

Peter

-- 
------------------------------------------------
Peter Rice, LION Bioscience Ltd, Cambridge, UK
peter.rice at uk.lionbioscience.com +44 1223 224723